Integrating microbiology and modelling to determine the source of Campylobacter infection in the broiler house and develop interventions

Campylobacter is the largest cause of recognised bacterial gastroenteritis in the developed world. The 2009 reporting rates for Great Britain show more than 64 000 cases, an increase of 30% in Scotland and 14% in England & Wales on the previous year, that has continued into 2010. Because there is substantial under-reporting of campylobacteriosis, the actual number of cases in 2009 is likely to be closer to 450 000. Further, about 10% of reported cases are hospitalised. This rise is all the more disappointing because rates of infection with Campylobacter had been falling between 2000 and 2005. Molecular strain typing, by us and others, has identified that poultry is significantly the most important source of this infection with the most common types found in human beings also being the most common in chickens. Studies on retail poultry show a prevalence of Campylobacter in this meat of over 65% with the main routes of infection being eating improperly cooked meat or cross-contamination to uncooked foods. To reduce this burden of human disease, action must be taken to reduce Campylobacter loads in poultry and The Food Standards Agency, Defra and BBSRC have all identified this as a major priority. The FSA is considering targets for the reduction in levels of Campylobacter in raw chicken at retail, to be achieved by April 2015. The target will be set and achieved through stakeholder engagement and partnership working. Interventions in the poultry industry abroad have resulted in dramatic decreases in human infection rates. For example, in Iceland where freezing of positive carcasses is used, in New Zealand where interventions and regulations were introduced and in the USA where improved hygiene and the use of chlorine washes for carcasses has been implemented. However, UK industry has largely been unable to achieve reductions. Although strategies such as poultry vaccination are attractive in the longer term, more immediately it will be through informed biosecurity interventions on broiler farms that control is likely to be most readily achieved. Indeed UK producers widely recognise that where robust biosecurity remains unbreached, as for the valuable (grand)parent birds that are used to produce the eggs that hatch into broilers, then Campylobacter colonisation is uncommon. It is in the high throughput broiler production that colonisation regularly occurs and where novel biosecurity controls, as proposed here, could play an important role. Our previous studies of the sources of Campylobacter infection in humans not only identified the principal source as broiler chickens, it also identified that the distribution of Campylobacter strains found in humans and in the reservoirs of chicken, cattle, sheep, wild birds, pigs etc, were quite distinct with some strains common to several hosts. This proposal seeks to better understand the relative importance of the potential sources of Campylobacter in broilers by using a modelling approach. The hypothesis is that some Campylobacter strains and some Campylobacter reservoirs are much more important than others in this process and that it is only by quantitating their relative importance and their interaction with each other that it will be possible to robustly identify the sources of Campylobacter in the broiler house and hence introduce effective measures to prevent the colonisation of these birds during production. The findings will enable policy to be developed (e.g. code of practice) to define which control measures are most effective in keeping broiler houses Campylobacter free. This will strongly influence industry through improved farming practice.

Poultry is the most important source of campylobacteriosis and over two-thirds of chicken at retail are contaminated by Campylobacter. A reduction in the Campylobacter loads and ideally a reduction in their prevalence in poultry at production would be the most effective measure for reducing human disease incidence. However to date, biosecurity has not proved to be entirely successful and it is thought that this may be due to the diversity of reservoirs that Campylobacter resides in. Our, and others, work has indicated that some Campylobacter strains and some Campylobacter reservoirs are probably much more important than others in this process and that it is only by quantitating their relative importance and their interaction with each other that it will be possible to robustly identify the sources of Campylobacter in the broiler house and hence introduce effective measures to prevent colonisation. It will be parameterised by data collected in the study. Up to eight Campylobacter strains will be selected from the top strains isolated from human cases and from chicken, cattle, sheep, pigs and wild birds. The efficacy of different strains to move from reservoir species (cattle, sheep, pigeon, chicken) to chickens will be quantitated to determine the dose response of each strain. The environmental survival of the Campylobacter strains will be estimated in a range of matrices associated with the broiler environment (faeces from cattle, sheep, pigeons and chickens; soil; water) and survival tested at two temperatures (4C and 15C) over a period of up to 3 months. This proposal will use an agent-based modelling approach, which identify how macroscale dynamics emerge from microscale interactions, to simulate the transmission of Campylobacter from the environment to the broiler house and potentially back to the environment on real farms. This will be used to quantify which interventions would be most likely to reduce broiler colonisation.

All three investigators at UoA have built up contacts with the broiler industry and with policy makers. This includes an ESRC funded RELU study looking at the impact of the gastro-intestinal pathogen E.coli O157 on rural communities and economies and more recently an MRC /NERC /BBSRC /ESRC funded ESEI study, gip_net, which aims to develop a new paradigm for the understanding and control of gastro-intestinal diseases (including campylobacteriosis). Both of these studies have extensively engaged academic, farming, environmental, health care and public communities. Dr Sparks also has extensive contacts within the poultry industry. A reduction in the burden of campylobacteriosis is a declared main priority for BBSRC, Defra and Food Standards Agency, and by engaging with these players throughout the course of the study we anticipate disseminating findings as broadly as possible. This research aims to determine whether the simple and practical removal of animals or birds from the immediate vicinity of broiler houses will reduce broiler colonisation by the bacterial pathogen Campylobacter, since these same strains are routinely isolated from retail chicken and are predominant in causing human infection. Society. Campylobacter is the most frequently recognised cause of gastrointestinal illness both in the UK and other developed countries. The 2009 reporting rates for Great Britain show more than 64 000 cases, under-reporting suggests it is closer to 450 000, and rates are increasing. The cost of this has been estimated at £583M in 2008 in England and Wales. The majority of cases have symptoms that are not simply mild discomfort but are exceptionally debilitating and as such result in significant distress to both patient and relatives. There is an associated significant burden on the health care system both at GP level and from the 10% of hospitalisations. Further, there is extended loss of working days with a major impact on local and national economy. The widespread belief underpinned by scientific and medical case control studies are that the consumption or contact with contaminated broilers is the main infection route. A reduction in cases will both increase general health and improve quality of life. It would also reduce the burden on health services. Stakeholders in industry and policy. This proposal will identify what sources are the greatest risk for broiler colonisation. This will enable policy to be developed (e.g. code of practice) to define how far high risk reservoirs should be kept away from the broiler house. This will strongly influence industry through farming practice. It is essential to engage with industry throughout, to obtain permission to sample farms in the short term, but more importantly in the longer term to discuss the practical issues of farm management in relation to broiler colonisation. One of the leading UK broiler producers, VION Food Group Ltd, have been invited to attend progress meetings. We will engage with major policy makers. We have agreements with the FSA's Dr Mary Howell from the London office, and Drs Jacqui McElhiney and Jane Horne from Scotland and with Defra to discuss our findings and conclusions at three meetings over the course of the study. Academic. The project will further establish collaborations between three key Scottish institutions namely University of Aberdeen, Scottish Agricultural College and Moredun Research Institute. Micro-organisms from different stages of the study will also feed into separately funded proteomic research at the Moredun. All biological material will be held at UoA within the Campylobacter archive, which currently holds some 8,500 isolates - these are readily available for external study and are continuing to be sought by both industry and academic organizations. Dissemination of results will be via appropriate high IF journals and by submitting an article to the main UK trade journal Poultry World.